config.py

"""
Configuration Module
Contains all configuration constants for the Universal Translator application.
"""

# ============================================================================
# Model Configuration
# ============================================================================

class ModelConfig:
    """Configuration for translation models."""

    # Primary translation model
    MODEL_NAME = "swiss-ai/Apertus-70B-Instruct-2509"
    TITLE = "Universal Translator"

    # Model parameters
    DEFAULT_MAX_TOKENS = 1024
    DEFAULT_TEMPERATURE = 0.3
    MIN_TEMPERATURE = 0.0
    MAX_TEMPERATURE = 1.0

    # Temperature recommendations
    TEMP_LITERAL = 0.1  # More literal, word-for-word translations
    TEMP_BALANCED = 0.3  # Balanced (default)
    TEMP_CREATIVE = 0.7  # More creative, natural-sounding translations


# ============================================================================
# Language Configuration
# ============================================================================

class LanguageConfig:
    """Configuration for supported languages."""

    # Popular languages for the dropdown (display_name: language_code)
    # All languages supported by langdetect (55 languages)
    POPULAR_LANGUAGES = {
        "Afrikaans": "af",
        "Albanian": "sq",
        "Arabic": "ar",
        "Bengali": "bn",
        "Bulgarian": "bg",
        "Catalan": "ca",
        "Chinese (Simplified)": "zh-cn",
        "Chinese (Traditional)": "zh-tw",
        "Croatian": "hr",
        "Czech": "cs",
        "Danish": "da",
        "Dutch": "nl",
        "English": "en",
        "Estonian": "et",
        "Finnish": "fi",
        "French": "fr",
        "German": "de",
        "German (Swiss - Aargau)": "de-CH-AG",
        "German (Swiss - Basel)": "de-CH-BS",
        "German (Swiss - Bern)": "de-CH-BE",
        "German (Swiss - Graubünden)": "de-CH-GR",
        "German (Swiss - Lucerne)": "de-CH-LU",
        "German (Swiss - St. Gallen)": "de-CH-SG",
        "German (Swiss - Valais)": "de-CH-VS",
        "German (Swiss - Zürich)": "de-CH-ZH",
        "Greek": "el",
        "Gujarati": "gu",
        "Hebrew": "he",
        "Hindi": "hi",
        "Hungarian": "hu",
        "Indonesian": "id",
        "Italian": "it",
        "Japanese": "ja",
        "Kannada": "kn",
        "Korean": "ko",
        "Latvian": "lv",
        "Lithuanian": "lt",
        "Macedonian": "mk",
        "Malayalam": "ml",
        "Marathi": "mr",
        "Nepali": "ne",
        "Norwegian": "no",
        "Persian": "fa",
        "Polish": "pl",
        "Portuguese": "pt",
        "Punjabi": "pa",
        "Romanian": "ro",
        "Russian": "ru",
        "Slovak": "sk",
        "Slovenian": "sl",
        "Somali": "so",
        "Spanish": "es",
        "Swahili": "sw",
        "Swedish": "sv",
        "Tagalog": "tl",
        "Tamil": "ta",
        "Telugu": "te",
        "Thai": "th",
        "Turkish": "tr",
        "Ukrainian": "uk",
        "Urdu": "ur",
        "Vietnamese": "vi",
        "Welsh": "cy"
    }

    # Language code to full name mapping (for detection display)
    # Expanded to cover more languages that langdetect might identify
    LANGUAGE_NAMES = {
        "af": "Afrikaans",
        "ar": "Arabic",
        "bg": "Bulgarian",
        "bn": "Bengali",
        "ca": "Catalan",
        "cs": "Czech",
        "cy": "Welsh",
        "da": "Danish",
        "de": "German",
        "de-CH-AG": "German (Swiss - Aargau)",
        "de-CH-BE": "German (Swiss - Bern)",
        "de-CH-BS": "German (Swiss - Basel)",
        "de-CH-GR": "German (Swiss - Graubünden)",
        "de-CH-LU": "German (Swiss - Lucerne)",
        "de-CH-SG": "German (Swiss - St. Gallen)",
        "de-CH-VS": "German (Swiss - Valais)",
        "de-CH-ZH": "German (Swiss - Zürich)",
        "el": "Greek",
        "en": "English",
        "es": "Spanish",
        "et": "Estonian",
        "fa": "Persian",
        "fi": "Finnish",
        "fr": "French",
        "gu": "Gujarati",
        "he": "Hebrew",
        "hi": "Hindi",
        "hr": "Croatian",
        "hu": "Hungarian",
        "id": "Indonesian",
        "it": "Italian",
        "ja": "Japanese",
        "kn": "Kannada",
        "ko": "Korean",
        "lt": "Lithuanian",
        "lv": "Latvian",
        "mk": "Macedonian",
        "ml": "Malayalam",
        "mr": "Marathi",
        "ne": "Nepali",
        "nl": "Dutch",
        "no": "Norwegian",
        "pa": "Punjabi",
        "pl": "Polish",
        "pt": "Portuguese",
        "ro": "Romanian",
        "ru": "Russian",
        "sk": "Slovak",
        "sl": "Slovenian",
        "so": "Somali",
        "sq": "Albanian",
        "sv": "Swedish",
        "sw": "Swahili",
        "ta": "Tamil",
        "te": "Telugu",
        "th": "Thai",
        "tl": "Tagalog",
        "tr": "Turkish",
        "uk": "Ukrainian",
        "ur": "Urdu",
        "vi": "Vietnamese",
        "zh-cn": "Chinese (Simplified)",
        "zh-tw": "Chinese (Traditional)"
    }

    # Default target language
    DEFAULT_TARGET_LANGUAGE = "Spanish"


# ============================================================================
# Voice Configuration
# ============================================================================

class VoiceConfig:
    """Configuration for speech-to-text and text-to-speech."""

    # Default providers
    # Note: For faster response times, consider "Local Whisper (Tiny)" or "Local Whisper (Base)"
    # OpenAI Whisper API is more accurate but has network latency
    # Using gTTS as default for HuggingFace Spaces compatibility (no asyncio issues)
    DEFAULT_STT_PROVIDER = "OpenAI Whisper API"
    DEFAULT_TTS_PROVIDER = "OpenAI TTS"
    DEFAULT_TTS_VOICE = "nova"  # Must be valid for DEFAULT_TTS_PROVIDER

    # Voice output enabled by default
    DEFAULT_VOICE_OUTPUT_ENABLED = True

    # Performance optimization settings
    # Set to True to prioritize speed over accuracy (uses smaller models)
    FAST_MODE = False  # When True, switches to faster STT/TTS options

    # Note: Voice choices are dynamically loaded based on selected TTS provider
    # See voice_handler.py for provider-specific voice lists:
    # - OpenAI TTS: alloy, echo, fable, onyx, nova, shimmer
    # - gTTS: default

    # OpenAI TTS voices (kept for reference, not used in UI)
    OPENAI_TTS_VOICES = ["alloy", "echo", "fable", "onyx", "nova", "shimmer"]


# ============================================================================
# UI Configuration
# ============================================================================

class UIConfig:
    """Configuration for the Gradio UI."""

    # App metadata
    APP_TITLE = "Universal Translator"
    APP_DESCRIPTION = "Translate text or speech to any language with automatic source language detection"

    # Theme settings
    THEME_PRIMARY_HUE = "indigo"
    THEME_SECONDARY_HUE = "cyan"
    THEME_NEUTRAL_HUE = "slate"
    THEME_FONT = "Inter, system-ui, sans-serif"

    # Custom CSS
    CUSTOM_CSS = """
    .gradio-container {
        font-family: 'Inter', system-ui, sans-serif;
    }
    """

    # Input/Output dimensions
    INPUT_TEXT_LINES = 4
    OUTPUT_TEXT_LINES = 4

    # Slider configurations
    MAX_TOKENS_MIN = 128
    MAX_TOKENS_MAX = 2048
    MAX_TOKENS_STEP = 128

    TEMPERATURE_MIN = 0.0
    TEMPERATURE_MAX = 1.0
    TEMPERATURE_STEP = 0.1

    # Launch settings
    SHARE_LINK = False
    OPEN_IN_BROWSER = True


# ============================================================================
# Translation Prompts
# ============================================================================

class PromptConfig:
    """Configuration for translation prompts."""

    SYSTEM_PROMPT_TEMPLATE = """You are a professional translator. Your task is to translate the given text to {target_language}.
Provide ONLY the translation, without any explanations, notes, or additional text.
Maintain the tone, style, and meaning of the original text."""


# ============================================================================
# Error Messages
# ============================================================================

class ErrorMessages:
    """Standard error messages."""

    NO_INPUT = "No input provided"
    NO_AUDIO_INPUT = "No audio input"
    TRANSCRIPTION_FAILED = "Transcription failed"
    TRANSLATION_ERROR = "Translation Error: {error}"
    TRANSCRIPTION_ERROR = "[Transcription Error: {error}]"
    LANGUAGE_DETECTION_FAILED = "Unknown"


# ============================================================================
# Helper Functions
# ============================================================================

def get_language_name(lang_code: str) -> str:
    """
    Get the full language name from a language code.

    Args:
        lang_code: Language code (e.g., 'en', 'es')

    Returns:
        Full language name or uppercase code if not found
    """
    return LanguageConfig.LANGUAGE_NAMES.get(lang_code, lang_code.upper())


def get_popular_languages_list() -> list[str]:
    """
    Get a list of popular language names for dropdowns.

    Returns:
        List of language display names
    """
    return list(LanguageConfig.POPULAR_LANGUAGES.keys())


def get_language_code(language_name: str) -> str:
    """
    Get the language code from a display name.

    Args:
        language_name: Display name (e.g., 'Spanish', 'French')

    Returns:
        Language code or the original name if not found
    """
    return LanguageConfig.POPULAR_LANGUAGES.get(language_name, language_name.lower())